We will study both afferent and efferent synaptic mechanisms with the long-term objective of understanding the distinctive contributions made by type I and type II hair cells and by efferent control to vestibular function. Work is done in the turtle posterior crista and utricular macula. There are two specific aims: 1) To study afferent synaptic transmission by recording from afferent terminals and hair cells. Transmis- sion involves both quantal and non-quantal mechanisms with the latter being quantitatively more important in calyx-bearing, than in bouton, fibers. We will determine whether quantal activity in calyx-bearing fibers origi- nates in type I hair cells. The mechanisms of non-quantal transmission and its role in vestibular transduction will be assessed. Quantal activity in the two kinds of fibers will be compared in terms of the participation of postsynaptie NMDA and non-NMDA receptors, desensitization, glutamate transporters, presynaptic glutamate receptors, and retrograde transmission from afferents to hair cells. Morphological studies will clarify the distri- bution and number of ribbon synapses, as well as the identity and distribution of glutamate receptors and trans- porters. 2) To study the synaptic activity evoked by electrical activation of efferent fibers. Efferent responses of bouton units depend on their neuroepithelial location in the crista, ranging from weak excitation near the planum to strong inhibition near the torus. Calyx-bearing units show a strong excitation, consisting of both fast and slow responses. We propose to continue studies of synaptic and receptor mechanisms for each of the several classes of afferents. We will examine whether the subsynaptic cistern, a common feature of hair-cell efferent synapses, acts as a Ca 2+store that can modify efferent actions and determine how responses to sensory inputs in various afferents are modified by efferent activation. Morphological studies will clarify the distribution and numbers of efferent synapses, as well as the identity and distribution of nicotinic and other efferent-related receptors. Finding the distinctive contribution of type I hair cells may be critical to understanding human vestibular function as these hair cells outnumber type II hair cells in man and are more vulnerable to pathological agents. The proposal takes advantage of the turtle to study problems that would be difficult, if not impossible, to study in mammals. One advantage is the variety of efferent actions that can be studied in a single organ.